Struct PolicyGateExecutor 

pub struct PolicyGateExecutor<T: ToolExecutor> { /* private fields */ }

Wraps an inner ToolExecutor, evaluating PolicyEnforcer before delegating.

Policy is only applied to execute_tool_call / execute_tool_call_confirmed. Legacy execute / execute_confirmed bypass policy — see CRIT-03 note above.

Implementations

impl<T: ToolExecutor> PolicyGateExecutor<T>

pub fn new( inner: T, enforcer: Arc<PolicyEnforcer>, context: Arc<RwLock<PolicyContext>>, ) -> Self

Create a new PolicyGateExecutor.

pub fn with_audit(self, audit: Arc<AuditLogger>) -> Self

Attach an audit logger to record every policy decision.

pub fn update_context(&self, new_ctx: PolicyContext)

Write the current context (called by the agent loop when trust level changes).

Trait Implementations

impl<T: ToolExecutor + Debug> Debug for PolicyGateExecutor<T>

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<T: ToolExecutor> ToolExecutor for PolicyGateExecutor<T>

async fn execute( &self, _response: &str, ) -> Result<Option<ToolOutput>, ToolError>

async fn execute_confirmed( &self, _response: &str, ) -> Result<Option<ToolOutput>, ToolError>

Execute bypassing confirmation checks (called after user approves). Default: delegates to execute.

fn tool_definitions(&self) -> Vec<ToolDef>

Return tool definitions this executor can handle.

async fn execute_tool_call( &self, call: &ToolCall, ) -> Result<Option<ToolOutput>, ToolError>

Execute a structured tool call. Returns None if tool_id is not handled.

async fn execute_tool_call_confirmed( &self, call: &ToolCall, ) -> Result<Option<ToolOutput>, ToolError>

Execute a structured tool call bypassing confirmation checks.

fn set_skill_env(&self, env: Option<HashMap<String, String>>)

Inject environment variables for the currently active skill. No-op by default.

fn set_effective_trust(&self, level: SkillTrustLevel)

Set the effective trust level for the currently active skill. No-op by default.

fn is_tool_retryable(&self, tool_id: &str) -> bool

Whether the executor can safely retry this tool call on a transient error.